Citrus fruits are the second most important crop worldwide. One of the most important tasks is sorting, which involves manually separating the fruit based on its degree of maturity, and in many cases, involves a task carried out manually by human operators. A machine vision-based citrus sorting system can replace labor work for the inspection of fruit sorting. This article proposes a vision system for citrus fruit sorting implemented on a dedicated and efficient Field Programmable Gate Array (FPGA) hardware architecture coupled with a mechanical sorting machine, where the FPGA performs fruit segmentation and color and size classification. We trained a decision tree (DT) using a balanced dataset of reference images to perform pixel classification. We evaluate the segmentation task using a pixel accuracy metric, defined as the ratio between correctly segmented pixels produced by a DT and the total pixels in the reference image segmented offline using Otsu’s thresholding algorithm. The balance between correctly classified images by color or size and their corresponding labels of that color and size evaluates the color and size classification algorithms. Considering these metrics, the system reaches an accuracy of 97% for fruit segmentation, 94% for color classification, and 90% for size classification, running at 60 fps.

This conlribution deals with the construction of a cartography of the internal surface of organs based on image sequences acquired during clinical video-endoscopic exams. This paper describes a metrological approach to adapt the images acquisition conditions and the pre-processing tools to the evaluation of our specifically developed registration and mosaicing algorithm. A 3D-micrometric positioning system and an endoscope's tip holding structure was used to acquire sequences of images following a specific protocol. Forty images of a test scene (bladder planar photography) were acquired along a determined displacement looped path. An algorithm for correcting non-linear radial distortion caused by the endoscope was implemented. This algorithm computes projective (camera) and polynomial (distortion) transformations. The optimization process regislers the corrected distorted pattern image with the non-distorted one. Mutual information was used as a measure of similarity and stochastic gradient descent method for optimization. The visual quality of the images was then improved by a shading correction and by removing the optical fibers pattern from the images (specific low-pass filtering). A low-pass filter was used for obtaining the background that is subtracted from every image. The registration algorithm based on mutual inrormation as similarity measure and stochastic gradient descent as optimization method was used to obtain the transformation parameters that were applied to the images for building a mosaic (cartography). Finally, these parameters were applied to a similar sequence of images of a dots pattern (in place of the photography) allowing us to compute significant error parameters for the quality of the mosaicing. The results of the various processes are presented and the evaluation of the performancc of the cartography is discussed.

This work proposes the evaluation of a set of algorithms of machine learning and the selection of the most appropriate one for the classification of segmented chromosomes images acquired using the Giemsa staining technique (G-banding). The evaluation and selection of the best classification algorithms was carried out over a dataset of 119 Q-banding chromosomes images, and the obtained results were then applied to a dataset of 24 G-band chromosomes images, manually classified by an expert of the Laboratory of Cytogenetic of the Children’s Hospital of Tamaulipas. The results of evaluation of 51 classifiers yielded that the best classification accuracy for the selected features was obtained by a backpropagation neural network. One of the main contributions of this study is the proposal of a two-stage classification scheme based on the best classifier found by the initial evaluation. In stage 1, chromosome images are classified into three major groups. In stage 2, the output of phase 1 is used as the input of a multiclass classifier. Using this scheme, 82% of the IGB bank samples and 88% of the samples of a bank of images obtained with a Q-band available in the literature consisting of 119 chromosome studies were successfully classified. The proposed work is a part of an desktop application that allows cytogeneticist to automatically generate cytogenetic reports.

The large-scale evaluation of multiple-choice tests is a challenging task from the perspective of image processing. A typical instrument is a multiple-choice question test that employs an answer sheet with circles or squares. Once students have finished the test, the answer sheets are digitized and sent to a processing center for scoring. Operators compute each exam score manually, but this task requires considerable time. While it is true that mature algorithms exist for detecting circles under controlled conditions, they may fail in real-life applications, even when using controlled conditions for image acquisition of the answer sheets. This paper proposes a desktop application for optical mark recognition (OMR) on the scanned multiple-choice question (MCQ) test answer sheets. First, we compiled a set of answer sheet images corresponding to 6029 exams (totaling 564,040 four-option answers) applied in 2024 in Tamaulipas, Mexico. Subsequently, we developed an image-processing module that extracts answers from the answer sheets and an interface for operators to perform analysis by selecting the folder containing the exams and generating results in a tabulated format. We evaluated the image-processing module, achieving a percentage of 96.15% of exams graded without error and 99.95% of 4-option answers classified correctly. We obtained these percentages by comparing the answers generated through our system with those generated by human operators, who took an average of 2 min to produce the answers for a single answer sheet, while the automated version took an average of 1.04 s.

Background: Automatic adventitious lung sound classification using deep learning is a promising strategy for objective respiratory disease screening. Evaluating model performance is challenging, particularly with imbalanced clinical datasets. This study compares CNN architectures and proposes a dual-stream classification approach. Methods: Using the public ICBHI 2017 dataset, we compared five pre-trained architectures: VGG16, VGG19, InceptionV3, MobileNetV2, and ResNet152V2. To mitigate class imbalance, we implemented pitch shifting, random shifting, and mixup data augmentation. We also developed and evaluated a novel VGGish-dual-stream network. The primary endpoint was the Average Score (AS), the arithmetic mean of Sensitivity and Specificity. Results: Among benchmarked models, ResNet152V2 achieved the highest AS ($0.541$), approaching the state-of-the-art range (0.56–0.58). This performance was characterised by a high Specificity ($0.67$) but low Sensitivity ($0.41$). Our proposed dual-stream network yielded a more balanced, albeit slightly lower, performance with an AS of $0.508$. Conclusions: Standard CNN architectures like ResNet152V2 can achieve competitive classification performance but may exhibit a clinically significant bias towards high specificity at the expense of sensitivity. This trade-off poses a risk of missing pathological events (false negatives). To ensure clinical safety and utility, future work must prioritise strategies that explicitly improve model sensitivity.